Towards a Human-Centred Artificial Intelligence Maturity Model

Abstract

Artificial intelligence (AI) is becoming a central building block of computational systems. Following the long traditions of human-centered design, Human-Centered AI (HCAI) emphasises the importance of putting humans and various societal considerations in the centre of the development. However, the question is: how to realise HCAI when designing systems that utilise novel computational tools and require consideration of increasingly broad set of requirements, spanning from fairness and transparency to accountability and ethics? The purpose of our study is to support the AI development practices in companies in order for the humans to have AI solutions that are efficient, trustworthy, and safe. To this end, we propose a maturity model for HCAI (HCAI-MM). In this paper we present the first phase of the model development, in which the central building blocks of HCAI are specified and initial company requirements for the model's structure and content are evaluated with four AI developers.